利用多尺度分析及压缩感知的图像融合算法研究.pdf

摘要
摘要
本文研究了多尺度分析的图像融合以及基于压缩感知的图像融合问题。
本文针对多尺度分析的图像融合问题，提出了三种基于多尺度分析的图像融合
方法。通过这三种方法，将两幅图像高频信息和低频信息进行融合。这三种方法
分别是：基于区域标准差的图像融合方法、基于区域熵的图像融合方法和基于区
域能量的图像融合方法。实验结果证明三种方法能够有效地提高图像的信息熵和
平均梯度。
本文针对压缩感知的图像融合问题，提出了基于压缩感知流程的两点改进方
法。实验结果表明，提出的改进方法与梯度取大的图像融合方法相比提升了融合
图像的质量，增强了压缩感知的性能。
本文提出的方法在解决多尺度分析的图像融合问题和压缩感知的图像融合问
题的同时，提升了图像的质量，有效地保留了图像的细节信息。

关键词：图像融合 压缩感知 多尺度分析
Abstract
Abstract
Image fusion method using multi-scale analysis and compress sensing is discussed
in this paper.
In order to solve the problem of image fusion using multi-scale analysis, three
algorithms of multi-scale analysis in image fusion are proposed in this paper. Through
these three algorithms, high frequency part and low frequency part are fused into a
single. The three methods are: image fusion method based on regional variance, image
fusion method based on regional entropy and image fusion method based on regional
energy. Experiment result shows that these three algorithms can improve both
information entropy and average gradient with effect.
In order to solve the problem of image fusion based on compressed sensing, two
improved algorithms are proposed in this paper. Experiment results show that the
proposed algorithms can improve both the quality of image and performance of
compressed sensing compared to the algorithm of reserving the large part that of greater
gradient.
These algorithms proposed in the paper can solve the problem in image fusion
using multi-scale analysis and compressed sensing, as well as improve the quality and
detail information of the image.

Keywords: Image Fusion Compressed Sensing Multi-scale Analysis
目录
目录
第一章 绪论 .........................................................................................................